Abstract
This paper deals with the estimation of travel time values for non-business inter-city passengers. Estimates are based on repeated survey data. A random coefficient probit model, which allows for individual heterogeneity, is shown to be slightly more successful with respect to prediction than an independent probit model, which is why the former is employed in calculation of time values. The sample is split into vacation and non-vacation periods, which can be seen as proxies for the length of the vacation. A considerable difference in time values is found in the two periods. Hence, when estimating the time values, in both stated and revealed preference studies for various transportation projects, it is important to consider the length of the vacation.
Get full access to this article
View all access options for this article.